| This dissertation describes the gender, race/ethnicity, and nativity patterning of undergraduate major choice and investigates the mechanisms leading to such distributions. To highlight the impact of pre-college experience, I focus on the initial choice of college major. Using a sociological perspective, I develop an integrated framework that underscores the role played by family, school, neighborhood and occupational structure in influencing college major choice. These institutional effects are both direct and indirect through pre-college experiences, including achievements, attitudes and course taking in math and English, which are identified as determinants of college major choice in the economic, psychological and educational literature.; I use data on individual, family, and school from the NELS:88-94 and construct measures of the characteristics and composition of occupational structure from the 5% PUMS of the 1990 census. The two phases of analysis include an examination of the effects of multiple institutions on pre-college experiences using panel data analysis, and an examination of the role of multiple institutions in influencing college major choice using multinomial logistic regression. The analyses correct for the selection bias resulting from nonrandom selection of students into college.; The analyses reveal that there are strong gender and racial/ethnic patterns in initial college major choice, but no significant differences between immigrant and native-born students. Women are still severely under-represented in technical majors and lucrative fields, albeit their lead in life/health fields. Math attitude is the only aspect of pre-college experiences under this study that partially accounts for gender differences in college major choice. I also find that non-Asian racial minorities, Afro-American students in particular, are not at all under-represented in technical majors or lucrative major fields. Minority students' over-representation in S&E for the first major and under-representation for the final major, as evidenced by previous studies, suggests "leaks" in the science and engineering pipeline. Another finding is about the contextual influences of family, neighborhood and societal occupational structure on college major choice. In particular, proportionate representations of demographic groups in occupational fields have significant positive influences on choosing the corresponding fields. This supports the close affinity between college major and occupation. The implications for theory and policy are discussed in the conclusion chapter. |
